Epidemiologic studies of the relationship of polyhalogenated biphenyl (PHB) exposure and adverse reproductive outcomes in humans have yielded conflicting results. The proposed study is a historical cohort design which will evaluate 741 women of childbearing age from two well characterized, highly exposed cohorts in Michigan. The "PBB" cohort and the "Silo" cohort were initiated 15 years ago by the Michigan Department of Public Health following exposure to PBB (through contamination of cattle feed) and PCB (via a PCB-laden sealant used in concrete silos). Data collected on these cohorts include medical and reproductive histories, behavioral characteristics, dietary histories of Great Lakes fish consumption (a potential source of PCB), PBB and PCB levels of farm products and buildings, and serum PBB and PCB determinations. Serum PBB level for the PBB cohort range from 0-828 ppb, median 2 ppb. (A previous study found no samples with detectable PBB from a control group of 222 Iowa women). Serum PCB levels in the Silo cohort range from 5-300 ppb, median = 26 ppb, compared to a background level of 4-8 ppb in the general population. The Michigan birth certificate computer file will be searched for the records of all infants born to cohort women between 1974-1990. Three control infants, matched on mother's age, race, infant sex and date of birth, and other confounding variables, will also be identified for each of the "exposed" infants through the birth certificate file. A comparison of birthweight and gestational age distributions, rates of preterm delivery, low birthweight, and fetal growth retardation, infant mortality, and congenital anomalies identified at birth will be performed, both between the cohort infants and controls, and within the cohort according to gradient of exposure. The database resulting from this research will provide a starting point for future studies on the largest cohort of infants exposed to PHBs in utero, with a carefully matched set of controls.